Birth weight and body weight at various points during lifetime have being suggested to influence the risk of premenopausal breast cancer. A high birth weight has been associated with an increased risk of premenopausal breast cancer. More curiously, overweight and obesity during childhood and early adulthood have consistently been linked to a decreased risk of premenopausal breast cancer. The mechanisms underlying these associations have not been elucidated. How body weight and body mass at different stages in life may interrelate in their impact on breast cancer risk has not been sufficiently explored. The opposite direction of the association of birth weight on one hand and of childhood and early adolescent body mass on the other hand with premenopausal breast cancer risk is perplexing. However, evaluating breast cancer risk in relation to these anthropometric factors during lifetime is complicated because not only may birth weight, childhood weight, and body mass during adolescence and early adulthood be related, these anthropometric factors can also influence other variables, such as age at menarche, height, which may be related to both later body mass and the risk of breast cancer. Furthermore, biological pathways underlying the association between anthropometric factors during lifetime and premenopausal breast cancer risk are not well understood. In animal experiments, obese rodents with a defect in the leptin receptor did not develop mammary tumors. We propose to investigate premenopausal breast cancer risk in relation to birth weight, relative weight at age 5 and 10, body mass index (BMI) at age 18 and updated BMI in early adulthood through biannual follow-up from 1989 through 2005 in the Nurses' Health Study II (NHS II). We further propose to evaluate leptin plasma levels and genetic polymorphisms of the leptin gene and the leptin receptor gene in a nested case-control fashion. We propose to examine whether breast density as detected in a mammogram may be an time-dependent confounder or an effect modifier of the association between body mass and premenopausal breast cancer. This will be the first study taking a life-course approach in linking body mass with breast cancer incidence. It is also highly efficient since it borrows from questionnaire data, blood samples, and mammograms already collected for the NHS II. [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable] [unreadable]